Re: [PATCH] 2.5.7 acct.c oops

Bill Davidsen (davidsen@tmr.com)
Fri, 22 Mar 2002 10:25:50 -0500 (EST)


On Thu, 21 Mar 2002, Bob Miller wrote:

> On Thu, Mar 21, 2002 at 04:34:49PM -0500, Robert Love wrote:

> > While he could of wrapped the test dependent in #ifdef/#endif
> > CONFIG_SMP, the test really is not overly needed. It is more a result
> > of the previous code, which Bob Miller himself fixed up and then much
> > rewrote the locking for.
> >
> > Since he recently did the cleanup (and even added that BUG_ON) I trust
> > he knows if we can remove it.
> >
> > Robert Love
>
> As Robert has sermized... When I did the cleanup I debated whether to put
> the BUG_ON() in there in the first place. I used it as an extra bit of
> paranoia not knowing that it would not return correct results on a UP.

Okay, that's a reasonable decision. I'm all in favor of paranoia, any
problem at this point would be best detected early on, nut if you're
convinced that it's safe without it, I certainly can't argue, and wasn't
intending to initially.

Actually I don't think CONFIG_SMP would completely avoid problems, if I
read the code right it would still have a problem if you had an SMP kernel
running "nosmp" option at boot (or possibly booted on a uni, although I
haven't tried w/o 'nosmp').

Thanks for all the information, clearly if any check is needed, that
check is not the one, at least as is.

-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/